The extraction of roads from satellite images is a necessary step for urban planning, intelligent transportation, etc. We propose Lite-HRNet-OCR, a lightweight and efficient CNN structure for road segmentation. The network of Lite-HRNet-OCR begins with a lightweight Lite-HRNet backbone that learns the weights of all channels and resolutions. The weights serve as the channel for information exchange across channels and resolutions. The multi-resolution output of the lightweight backbone is input to OCRNet, which organizes contextual pixels into object regions and exploits the relationships between pixels and object regions to augment the representation of their pixels. Two loss functions, cross-entropy loss and Tversky loss, are used to solve the problem of sample imbalance. Experimental results show that our method achieves competitive performance on the public CHN6-CUG road dataset. Specifically, the Lite-HRNet-OCR achieves 64.39% Mean IOU and 96.52% F1 with 2.9 MParams and 29.4 GFLOPs.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.